Introduction

According to Seabra et al. (2010), 94% of the liquid transportation fuels are derived from oil. Recently there has been an increased focus on bio-fuels due to the impact of fossil fuel consumption on global warming, which requires a reduction in greenhouse gas emissions, the increased energy demand worldwide and the depletion of fossil fuel reserves (Dias et al., 2011). Lignocellulosic biomass is a renewable resource ideal for energy production, as this sustainable feedstock can be converted to bioethanol by biological conversion.

First generation bioethanol is commercially established and is produced from feedstock such as corn and sugar cane however it is not sustainable as food crops are used as feedstock (Sims et al., 2010). Second generation ethanol is not commercially established but can be produced from sustainable (non-food crop) lignocellulosic biomass that include straw (Tomás-Pejó et al., 2008), sugarcane bagasse (da Silveira dos Santos et al., 2010), energy crops (Mishima et al., 2008) and various forestry and agriculture residues (Sims et al., 2010).

One of the main challenges of producing ethanol from lignocellulosic feedstock in general is that pre-treatment is required to increase the digestibility of cellulose and to maximise the sugar yield from hemi-cellulose (Cruz et al., 2011). However, most heat- and acid-based pre-treatment processes generate unwanted inhibitory by- products such as furans and phenolic compounds that negatively affects yeast

22

performance (Parawira and Tekere, 2011). Typically, such pre-treatment processes were carried out using steam explosion, dilute acid treatment, organosolv or sulphite pre-treatment (Zhu and Pan, 2010). Paper sludge, on the other hand, presents one major benefit in that no pre-treatment is required. This is due to the fact that significant disruption of the cellulosic crystalline structure occurs during the paper pulping process (Kang et al., 2010). As a result paper sludge is a viable feedstock for bioethanol production.

In the manufacturing of paper, pulp fibres are damaged and shortened, resulting in disposal of between 15 and 20% of the pulp feed stock as paper sludge during the manufacturing process (Jeffries and Scartman, 1999). The increasing population of South Africa will generate increasing amounts of paper waste in the future, making the conversion of waste a matter of increasing importance. Such conversion to ethanol, for example, would limit the amount of solid waste that needs to be disposed of (van Wyk, 2003), but could also serve as a valuable source of income through value addition. Paper sludge typically contain 50% or more carbohydrates with glucan and xylan as the main components (Lynd et al., 2001), making this material a suitable source for conversion to bioethanol.

The major drawback with second generation ethanol is that the cost of cellulase contributes significantly to the production cost of bioethanol and according to Klein- Marcuschamer et al. (2012) enzyme cost could be as high as $ 1.47 gal-1 (R 3.28 l-1) ethanol produced. For a SSF process to be economically viable at industrial conditions an ethanol concentration of 40 g l-1 is required to reduce product recovery cost (Fan, 2004). As a result it is critical to minimise cellulase dosage while optimising ethanol concentration and yield. Response surface methodology (RSM)

23

can be used to optimise cellulase dosage and paper sludge loading to obtain high ethanol concentrations and yields. RSM is a statistical technique comprising of a collection of mathematical and statistical techniques that can be used to model the response of a system influenced by several variables (Han et al., 2011).

Response surface methodology was used to model and optimise key aspects of SSF, namely enzymatic hydrolysis (Liu et al., 2009), pre-treatment of sugarcane bagasse (Cruz et al., 2011) and the fermentation of sago starch (Ratnam et al., 2003). The conditions optimised in the RSM with regards to fermentation of sago starch were time and temperature. In another study, RSM was used to optimise time, pH and temperature for the SSF of kitchen waste (Wang et al., 2008). However, RSM has not been used to date to optimise ethanol concentration and ethanol yield in SSF with regards to solid loading and cellulase dosage for any lignocellulosic feedstock.

In recent times there has been an increased focus on the use of paper sludge as a feedstock for the production of second generation ethanol using SSF. Most of this research was conducted on sludge emanating from the Kraft pulping process (Fan et al., 2003; Fan and Lynd, 2006; Kang et al., 2010; Zhang and Lynd, 2010). There is a lack of information available on the SSF process utilising recycled paper sludge as feedstock. Given this lack in information pertaining to an often neglected renewable source of energy, this study focused on determining the composition, fermentability and optimum solid loading and cellulase dosage for producing ethanol from recycled paper sludge.

The composition of the paper sludge samples was determined by the method prescribed by NREL and the fermentability of the samples was determined by SSF in

24

100 ml fermentation bottles. Based on ethanol titres from these experiments, two samples yielding the lowest and highest ethanol concentration were selected for optimisation by RSM in 1.3 L BioFlo Modular Benchtop Fermenters. In this work, RSM was used to maximise ethanol production and yield by optimising paper sludge solids loading and cellulase dosage. Multi response optimisation using a desirability function approach was used to optimise the results obtained from the RSM. The desirability function approach is widely used in industry to optimise a system with multiple responses (StatSoft, Inc, 2011). The data obtained from the optimisation of SSF will be used as input for process modelling and economic evaluation of ethanol production from paper sludge (Chapter 4).